Running Head: NATURALIZING ETHICS WITH COGNITVE SCIENCE

Naturalizing Ethics - Cognitive Science against Moral Relativism1

Sylvain Pronovost

Institute of Cognitive Science, Carleton University

1 Carleton University Cognitive Science Technical Report 2006-04. http://www.carleton.ca/ics/TechReports

1

1

Introduction

Ethics has for long been a privileged area of human activity untouched by

deterministic and materialistic claims as its particular subject matter is essentially

praxeological, that is the moral inclinations and conducts of people. With

respects to free will and self-determination, moral epistemology has undergone

significant changes over the centuries of occidental thought on moral judgments

and their particular normative aims. The problem is, under the circumstances of a

diversity of intellectual schemes and practices as observed worldwide by

scientifically inclined thinkers, establishing morality on strong foundations and

assuring a coherent scheme with metaphysical and epistemological

presumptions became apparently harder as time has gone by. This has led to a

polarization between two radical theories on the status of judgments of values, in

the diverse avatars of absolutism and relativism.

Gilbert Harman (1985), fascinated with the ongoing allegiances towards

either radical claims (and noting that a rare few were uncommitted) and with

apparently no accepted grounds for a settlement, has reoriented the issue on a

broader epistemological conflict between naturalism and autonomous ethics. My

observations have led me to agree with this argument, but also under an

orthogonal perspective, that is, the relativism and absolutism conflict generally

revolves respectively around a cultural and external account of the nature of

morality versus a rational and internal account of moral principles. On the first

account, ethics are immanent to social and cultural interactions, and morality

seems incommensurable, therefore voiding the possibility of there being

universal moral truths. On the absolutist side, ethics are (almost) transcendent, a

priori to human activity, and through rational endeavors it is possible to grasp or

develop universal and general principles of morality that are invariably true.

2

I hold that the debate on ethical relativism is reductive. It is too

preoccupied with the dual commitments of naturalization contra autonomy on one

level, and drawing upon it being externally (empirically) versus internally

(rationally) determined on another level, clouding a deeper disagreement in the

form of a misinterpretation of the very concept of relativism. I will therefore

dedicate the first part of this essay to a reinterpretation, or clarification, of what it

is for a belief or a set of beliefs to be held in a relativistic framework, inspired by

analogical empirical and formal systems of beliefs. In a second section, I will

argue against the possibility of an autonomous study of ethics, drawing upon the

debate on normative and descriptive clauses. I will also assess the coherence

and fallibility of an autonomous research program, notably by pointing out its

underlying metaphysical and epistemological commitments. In the following

section, I will focus on the opposite view, the naturalization of ethics, and how it

follows naturally from contemporary metaphysical and epistemological beliefs

and endeavors, aiming to adjust to preoccupations of psychological realism,

empirical adequacy and theoretical constraints. Finally, following a massive

quantity of research in the interdisciplinary sciences of cognition, I will offer an

account of what type of constraints and reliable sources of information

naturalized and relativistic ethics may draw upon in the pursuit of moral inquiries.

2

The concept of relativism under scrutiny

A definition of relativism

Let’s start with a relatively (no irony is intended here) uncontroversial

definition of what it is for a theory of knowledge to be dubbed of relativist. I have

chosen a “clean” definition that is neither too obscure, nor too narrow and

restrictive in my opinion, from Bénatouïl (1997), “… le relativisme consiste à

affirmer, dogmatiquement, que nos jugements et nos valeurs n’ont de sens que

3

relativement aux sujets et aux contextes qui les produisent, leur prétention à

l’objectivité et à l’universalité étant remplacée par une validité

psychologiquement, historiquement, ou socialement limitée. “

What is fascinating about relativism is the age-old claim (of Greek origin,

as this problem has been haunting at least the Occidental part of our world for

quite a while) that it is a paradoxical dead end, an undesirable logical

troublemaker to which we can not ascribe a truth value because its form (an

absolute claim) necessarily contradicts its content (about nothing being

absolute). The claim of relativism being, prima facie, paradoxical, is roughly as

follows:

All statements / theories are relative.

Following pop culture in its inexorable generalization, or what scholars

sometimes dub “folk science” or “naive science”, this already mind boggling

puzzle of logic as gone as far as

Everything is relative.

which is truly blown out of proportion. Now, one might be tempted to abandon the

relativistic outlook of knowledge for its hopeless tone, or either champion it or

rebuke it in this very narrow form (to my sense, which I will justify in what

follows). Literature is prolific about the “inadmissibility” or the “inevitability” of

relativism, from theoretical knowledge to praxeological knowledge, from physics

(even mathematics) to ethics. Let’s be honest: in this crude form, either relativism

dooms us all to the underdeterminacy of a referential framework, undermining

the very concept of truth and “mutual agreement” on any statements or course of

action in a world of caeteris paribus (all things being equal) or petitio principii

(begging the principle or motive), OR relativism is inadmissible, has to be

admonished and the alternative is to look to the stars in search of universals.

Well, I may have a romantic point of view of the ongoing debate, that is,

the one between relativists and absolutists, but they leave a bitter sense of

dichotomic, binary concepts when it comes to the study of knowledge. In both

camps, one ought to find some clause that will threaten to shake the edification

of its being a very radical thesis on the nature of knowledge. One good trick of

4

the philosophical trade, although it may be inadequate or misleading, is to

reassess the problem in its very definition. Conceptual analysis, as it is, may be

instrumental not only in solving problems of theoretical nature, but also in

assessing if a problem has been correctly formulated, or in a more extreme

measure, if it is really a problem at all. My claim here will be that relativism, along

with the formulation in which it is usually presented, are semantically wrong,

misleading us to other incorrect claims and edifying theories and counter-theories

that hold on a misinterpretation. I will also discuss in the next section on the

metaphysical and epistemological commitments that entail to such a

misinterpretation, which are radical claims about knowledge and the world that

are themselves controversial and questionable.

Anecdotes on relativism and relativity

Maybe a quick excursion in the realm of empirical and formal sciences

could teach us a lesson or two on what is meant by strong claims of relativism in

/ of a system. In mathematics, relative clauses are translated in concepts such as

dependent and independent variables, where a fixed magnitude and a dependent

one express real world phenomena. Mathematical descriptions using algebraic or

geometrical representations hold only through isomorphism (accurate mapping of

relations) or homomorphism (accurate mapping of relations and topological

properties of objects) in relation to the phenomenon they purport to describe.

Independent variables are set in a domain, which could be described as an

ontology of origin, and through a study of relations, most likely “strong” ones that

have a mapping of variables of one on one that we commonly name functions,

the end result is a description of a second ontology set in a co-domain, or usually

named an “image”. Of importance here is the non-causal, non-necessary, as

such arbitrary description of phenomena, but causal relations may also be

portrayed mathematically. The key element is that such systems are purely

descriptive, and only the actual phenomena are, to a certain extent, non arbitrary.

Also, within formal models (their internal structure), that a formal description

5

includes relative clauses is so up to a certain level of regression both in

mathematics and in logic, namely to the level of axioms. Hence, we may have yet

to progress with caution when we claim that formal systems, in their descriptive

role, are arbitrary. Internally, even formal sciences have an upper bound of

arbitrariness, and axioms are not relative clauses.

Yet another journey into relative claims or even relative systems of beliefs

has been available for the best part of the twentieth century in a sadly

misinterpreted theory of mythical proportions: the theory of relativity developed

by Einstein (1918) in its two avatars: general and special theories of relativity. In

an obviously (reductive) general and nonmathematical way, here is what it says:

“… in simplest terms, the theory of relativity is an approach to the

measurement and study of space and time. The theory assumes that findings

are based upon the relation of the frame of reference to the objects measured…”

“… both [theories] hold that certain physical quantities, formerly

considered objective, are actually ‘relative to’ the state of motion of the

observer.”

Now, history has demonstrated how shamefully misunderstood have been

Einstein’s claims about the nature of our knowledge of the physical universe.

Following the paradoxical interpretation of relative clauses within a system of

beliefs or even the relativity of wholly articulated systems of beliefs between

themselves, it is as if Einstein’s work had reinforced the naive conception of

relativism: everything is relative. In the realm of physics, people’s source of

inspiration for the reinforcement of this belief comes from a troubled

understanding of the following: observational statements about motion and

position of objects are dependent on the measure involved and the status of the

observer. Therefore, macrophysics being a set of descriptions of the behavior of

magnitudes dependent on one another, it’s all relative. Everything is relative.

Wrong. Naive / wrong physics lead to a larger set of naive / wrong beliefs

reinforcing the archaic relativist claim, just as its paradoxical formal structure is a

false problem which finds its source in an apparent ambiguity of the semantics of

relativism. We will explore my claim about the relativity of descriptions in the

6

empirical realm of physics first, and then the controversial claim that I just made

about the semantics of relativism.

I do not want to drown the reader in empirical details of highly technical

and abstract nature, but here is the proper treatment of the theory of relativity, as

reductive as possible:

“… when we have the above object moving, it will have a certain amount

of energy. Einstein argued, the only way we can insure that it cannot be

accelerated indefinitely, is if there is a universal equivalence between mass and

energy. The more energy an object has, the heavier it will be. When we speed it

up a little bit it becomes a bit heavier, and so it also becomes a bit harder to

speed it up further. In fact, the closer we are to the speed of light, the larger the

force is needed to accelerate the object; an infinite force is needed to speed up a

material object to the speed of light: it never happens!”

“… [on general relativity] Special relativity made the velocity of light a limit

for all causal processes and required revision of Newton’s theory of gravity as an

instantaneous action at a distance…”

“A philosophical motive for the general theory was to extend the relativity

of motion. Einstein saw special relativity’s restricted class of equivalent reference

frames as an ‘epistemological defect’, and he sought laws that would apply to

any frame… [after the physicalist justification] Thus not only velocity and rest, but

motion in general would be relative.”

“The curvature of space is real and is generated by the mass of the

bodies in it. Correspondingly the curvature of space determines the trajectories

of all bodies moving in it. The Einstein equations are the mathematical

embodiment of this idea. Their solutions predict, given the initial positions and

velocities of all bodies, their future relative positions and velocities. In the limit

where the energies are not too large and when the velocities are significantly

below c the predictions of Einstein's equations are indistinguishable from those

obtained using Newton's theory. At large speeds and/or energies significant

deviations occur, and Einstein's theory, not Newton's, describes the

observations.”

7

Conclusion: relativity is a strong claim about empirical descriptions

concerning macrophysics that misled many people to the wrong belief that these

descriptions belong to a relativistic theory of macrophysical phenomena, and

physics being a core discipline as it is, part of the “hard sciences” of nature, this

misconception is reinforced ad hominem in support of the archaic claim of

relativism. Here is what Ian Stewart (1990), mathematician at the University of

Warwick, says about the general treatment of the concept of relativity, “[following

a discussion on the concept of chaos in mathematics]... The same happened to

Einstein’s relativity theory which was widely used in the United States as an

excuse for social inequality. ‘Everything is relative, as Einstein says’, became the

chant. Not so. The most interesting thing that Einstein said is that some things,

notably the speed of light, are not relative.”

Relativism redefined and ethical significance

To maintain an unambiguous and rigorous semantics of what I purport to

clarify about relativism, here is a short version of relativism redefined:

a belief or system of beliefs is / are relative in nature with respect to an

invariant set of beliefs that are in turn justifiable, either internally or externally.

Even shorter: a claim or set or claims is / are relative to something else, they

are not relative in essence ! The following precisions must be brought up:

- something being relative to something else is so in virtue of formal,

methodological, theoretical or empirical reasons. Therefore, nothing in

relativism entails that relativity is so by itself.

- This relative dependence on an invariant core of belief is not a form of

absolutism: the set of invariants on which the relative clauses depend are

NOT a priori, they might themselves be scrutinized along a dependence

relation to other bodies of knowledge.

Some might be tempted to dub my conception of knowledge as holistic, where

beliefs are held in relation to others ad infinitum, in complex interdependence

where we have only access to our representations and phenomena, never the

8

factual basis of this knowledge. That is not what I believe to be so. My claim is

simply that in every body of knowledge, a set of core beliefs are taken as such as

postulates or axioms for the rest of the system. Their truth might be challenged

and we may turn up with better theories, but this is not a sufficient argument

against realism, nor is the thesis of the inscrutability of the reference.

My point can be resumed as follows: relativism is an interesting claim

about knowledge, but misunderstood in its very semantics. As many archaic

dichotomies have been progressively overthrown in favor of mediate solutions or

entirely different ones over the history of science, so has the false problem of

absolutism versus relativism got to give up. Perhaps we should think of it as a

spectrum, with two poles being a set of “justified true beliefs” on one end, and a

larger set of relative clauses that hold only in virtue of the former on the other

end. We will explore the archaic alternative views of knowledge on which the

modern theatre of relativism and absolutism clashes take place in the following

section, and make sense of the precedent claim about a reconstruction of the

concept of relativism in the context of ethics and its dependence on empirical

sciences. It will become apparent that the epistemological and metaphysical

foundations of the rift between absolute and relative knowledge are responsible

for the complications that have generated an even larger misconception, the

controversial normative and descriptive divide.

3

Moral epistemology on trial: autonomy or naturalization ?

A legacy of autonomous ethics, from Hume to Moore, not forgetting Kant…

We will follow Mark Johnson (1993, 1998) and Robert Audi (1998)’s

accounts of the epistemological debacle over ethics. Why is there so much

resistance from the study of ethics to delve into the realm of the empirical in

search of wisdom ? From the empiricist claims of Hume (1739) to the rationalist

9

convictions of Kant (1797), quite opposite views of the world, spun forth a great

epistemological gap over the status of empirical and moral claims. The is/ought

dichotomy appeared to Hume as quite a natural state of the world. In his view,

facts and values were totally independent of each other, and there was no

justification for moral philosophers to admit descriptive statements of factuals into

the realm of normative statements of values. Kant viewed the sphere of ethics as

an independent realm of moral laws, universals and necessary truths that were

available to the scrutiny of pure reason. Empirical lore was available through

synthetic and a posteriori scrutiny, whereas moral principles would be available

through the exercise of rationality alone.

But it probably was G.E. Moore (1903)’s position that undermined later

endeavors in being even interested in a possible reunion of the practical and the

empirical. His radical view held that moral concepts are non analyzable in

empirical terms and possess nonnatural properties of experience (by contrast

with analyzable and natural properties of objects and states of the world available

to perception and sensation). Johnson notes that Moore never really explained

what was meant by nonnatural properties. On Moore’s account, moral judgments

and explanations can not draw upon empirical references, such as in the

determination of the concept of what is good. He named the position of

attempting to relate descriptive references to the development of normative

clauses the position of naturalistic fallacy.

To sum up these views, moral philosophy ought to be about rational

analysis and normative claims, since values are of another nature than that of

facts. But what then, might I deservingly ask, are the tools and trade of moral

thinking ? If by reason alone, it aims to be rule- or law-like internal coherence.

Autonomous ethics then revolves around conceptual analysis, formal exercise

and evaluation of concise situations (or counter examples). It therefore seems

that ethics appeal to a rational, logical exercise of the mind in order to produce or

discover evidential truths of universal, normative essence. This is a bold gesture,

but before we follow up on the other side of the normative-descriptive divide, that

is in the realm of the dependency of ethics on empirical matters, we may want to

1

explore a certain number of difficulties that are internal to such a claim of the

autonomy of a rational view of ethical deliberation.

On the use of logic

The twentieth century dealt with some lethal attacks on the workings of

deductive logic, following some mathematical and logical theorems that

successfully repudiated the apparent universality of logical systems such as first-

order predicate logic and all of the non classical deployments of propositional

calculus, such as modal logic and polyvalent systems. The pretense of

universality and internal coherence of these avatars of logic was shattered with

theses on incompleteness and undecidability within formal systems (Gödel 1931,

Church 1936). Whether these considerations have any bearing on moral

philosophy should be carefully weighted, since the use of formal-deductive

judgments involving values rarely reduce to classical propositional calculus,

mostly calling upon modal systems. Logic is an articulate representation of the

internal coherence of arguments, and as such is really helpful for human

knowledge, but it is fallible.

An epistemological prejudice

Rationalist accounts of moral philosophy take many things for granted

about how we relate to the world. It is entirely up to debate (and quite

controversial anyway) as to whether there is such transparence of reason and

freedom from external constraints on this same reason for the moral agent. The

autonomy of ethics also presupposes an internalistic account of the

comprehension of moral principles, since these are either developed by the

exercise of reason, or worse, they are “discovered”, floating around and available

to the wise. Autonomous ethics also take verifunctionality for granted, in a

preference for meaning realism with a weak conception of justification such as

would be challenged by Michael Dummett (1993), although outside of ethics. To

1

my sense, justification of moral judgments, in order to claim independence from

the dominion of empiricism, has to have an aprioristic account of justification.

A metaphysically radical foundation

Autonomous ethics, as I claimed above, are based on a realist account of

moral principles. Whether they are constructed or discovered, moral truths or

moral laws possess a strong ontology, where principles bear a univocal,

universal and quasi- to literally transcendent status. A priori moral principles

seem to come before moral practice and hold whether or not reason follows

them. One last point worth noting is the teleological essentialism that permeates

autonomous moral thought. If moral principles are “available” to human reason,

whether created or discovered, they have this external teleological appeal to

them that is taken for granted. It may be that moral truths or principles are

teleological, but not necessarily by their being autonomously available through

ethical exercise of reason.

While the reader might object to these considerations about the internal fallibility

of autonomous ethics (and rightly so, since I only mean to raise properties of

internal coherence of autonomous ethics to critical scrutiny), it is in no way

preventing me from moving on to my more extensive thesis on the naturalization

of moral epistemology.

4

From naturalized epistemology to naturalized ethics

Saving something from Mill, visiting Dewey

Granted, utilitarianism did not fare well, it was doomed by many

ambiguities and its idealistic presuppositions of the likes of what would be a

proper definition of good (1861). But there’s something about Mill: his defense of

1

external determination of moral judgments. Externalism is very important for the

naturalistic enthusiast. After so much pressure from the naturalistic fallacy in the

making of modern moral philosophy, moral psychology did not have much to say

about normativity. John Dewey (1922) tried his best to bring back empirical

science into the realm of moral philosophy, to the extent of claiming that

empiricism was the core of ethics. His point was that since ethics concern human

nature directly, empirical claims on the body, mind and social organization of

moral agents are not only pertinent but indeed essential to moral exploration.

But according to Johnson (1998), Dewey was ignored by Anglo-Saxon

analytic philosophy of morals. In the later half of the twentieth century though, as

empiricism and naturalistic epistemology have taken over the sciences of the

individual and the social, moral theorists have grown to be even more concerned

with the normative-descriptive dichotomy.

Flanagan’s concerns with psychological realism

The tide as turned, and empiricism is back in the theatre of morals. The

later half of the twentieth century saw the birth and rise of interdisciplinary fields,

where sciences met to further global and specific research programs by the

means of collaboration. One dominant current as been that of the cognitive

sciences, from their early endeavors under the label of cybernetics to the now

immensely prolific areas of artificial and biological intelligence. But cognitive

sciences are not only a diversity of areas of research, they are also a set of

epistemological and metaphysical claims, a variety of methodologies, and an

encompassing thesis over the life sciences and the sciences of individuals and

societies: the thesis that all biological, and certain artificial entities, from their

inner workings to their associations, can be heuristically described and explained

through the concepts of information processing, intelligence, and the like.

Owen Flanagan (1991) has voiced an elementary concern to the study of

ethics, in light of the importance of the cognitivist approach in contemporary

empirical research programs, that he formulated as “… the principle of Minimal

1

Psychological Realism: make sure when constructing a moral theory or

projecting a moral ideal that the character, decision processing, and behavior

prescribed are possible, or are perceived to be possible, for creatures like us”

(my emphasis in italics). As Johnson notes, the principle may sound “innocuous”,

but its implications and consequences are radical in the context of ethics. It

places normative ethics in a direct, and necessary relationship with what we

know about the mind, and behavior of moral agents. And that is precisely what

cognitive sciences purports to study. The philosophical “prejudice” against the

descriptive and empirical cognitive sciences is here under trial, and supporters of

autonomous ethics will here demand justification. Well, I will try to demonstrate,

with the help of philosophers and cognitive scientists who have had a say in the

relevance of their endeavors to the field of ethics, that the evidence supporting

the claim of the necessity of grounding ethics in cognition is not only convincing,

it is a lethal blow to any pretense of the autonomy of any area of knowledge

about human life. And then, I will proceed to demonstrate the relativity of ethics

not exclusively to culture (as anthropologists would have it, like Benedict, 1934),

but to a larger scheme of constraints, from the biological to the social, under the

guidance of cognitive sciences’ results.

Naturalistic endeavors and the autonomous ethics fallacy

Johnson summarizes the importance of mentalistic issues in moral

epistemology, “… first, any plausible moral system must be based on reasonable

assumptions about the nature of concepts, reasoning, and moral psychology.

Second, the more we know about such important issues as the role of emotion in

moral deliberation, the nature of moral development, and the most realistic

conceptions of human well-being, the more informed we will be in our moral

thinking.” He then takes on utilitarianism and kantianism to prove his point:

utilitarianism presupposed highly determinate concepts such as that of “good”,

and capacities of rational calculation over the realization of what is good for self-

and communal purposes. Kantianism held absolute, universal moral laws, under

1

the guidance of a transparent and all-encompassing reason, where all of these

concepts were taken to be literal, clearly defined and available to all “sufficiently”

rational agents.

Those theories are psychologically unrealistic. They are doomed to the

status of idealistic outlooks on what normative morals should be. We have

already discussed the reasons for the origins of the normative and descriptive rift,

and it will become more and more apparent, as we progress, that maybe the

denial of the relevance of empirical research on cognition was a bad move for

ethics. So, where do we start ? Johnson says that “from the perspective of the

cognitive sciences, then, the most fundamental challenge is to show that the

alleged is/ought split is mistaken…” I have dubbed the departure from

autonomous ethics the “autonomous ethics fallacy”, in a (admittedly sarcastic)

tribute to Moore.

The unbearable lightness of language, where words betray metaphysics

A fundamental issue in the following discussion, as well as a clarification

of the reworking of the definition of relativism above, revolves around the

metaphysically “charged” concept of criteria. From ancient to modern philosophy

and sciences, the concept of criterion has been a terrible metaphysical

commitment in what constitutes knowledge and ontology alike. It presupposes

much in the foundations of epistemology and ontology, and efforts have been

made at the beginning of the twentieth century by logical positivists to “clean up”

the linguistic pitfalls and biases that committed many thinkers into metaphysical

delusions by putting too much “faith” in their conceptual schemes (Carnap,

1928). Quine (1953, 1990) did stop the frantic logico-positivist agenda from

exhausting its radical pogrom, but also warned his posterity of the dangers of the

underdeterminacy of conceptual translation and the inscrutability of ontology.

Thinkers, he warned, should be careful about their claims on what there is, thus

enforcing a strong position of ontological commitment in our use of language.

1

My point about criteria is that it’s overrated, out (of fashion) and obsolete.

Only formal systems may carry such a strong commitment on their design, by

their purely artificial and descriptive nature and function. Empirical sciences, up

to and including economics, have been influenced by many thinkers concerned

with questions on teleology (Dawkins 1986, Dennett 1995, 2003, Millikan 1987),

complexity and determinism (Lewin 1993, Coveney & Highfield 1995, Stewart

1990, Gleick 1988, Kauffman 1993) and rule-based systems (Nash 1997,

Axelrod 1984, Simon 1956, Newell 1990). These endeavors have changed the

face of the epistemology of science as well as its metaphysical foundations, and

contemporary research from even mathematics to physics to all of the life

sciences have departed from the metaphysically charged concept of criterion to

embrace that of constraint.The concept of constraint has a lighter touch in terms of metaphysical

commitment. It doesn’t place us in a world of ontological density (where

everything from ideas to objects that you can think of are granted reality a priori),

but rather in one of ontological propensity (where there may been something out

there, from concepts to abstract entities to material ones). Note how it is also

epistemologically lighter, in its “prudent realism”, for we are committed to the

limitations of not only ontology, but above all, our knowledge of the world. Our

claims are weaker with constraints, but at least they do not commit us to idealistic

schemes of ill-fated grandeur, of the likes of kantianism and utilitarianism. As

such, the cognitive sciences-inspired individuals have a better understanding of

what Johnson means when he writes “…we must examine the ways in which

cognitive science constrains moral theory…”, and later “… for the most part,

results from the cognitive sciences available to date function primarily to set

constraints on the nature of a psychologically and cognitively realistic morality…”

(my emphasis in both quotes).

1

5

What cognitive sciences have to say about ethics

On concepts, rules/laws and reasoning

Moral law has been a dominant ethical model from ancient Greece to

religious thought, to absolutist views of the like of intuitionism, rationalism and

even rawlsian judicial thought. “… Morality is regarded as a system of universal

moral laws or rules, discernible by human reason, and directly applicable to the

kinds of concrete moral situations that people encounter in their lives…” writes

Johnson (1993). He then emphasizes that cognitive research on

conceptualization and reasoning is of direct consequence to the viability of moral

law theories. These theories hold on premises such as universality and explicit

rules, and are incompatible with cultural and historical relativism. Drawing upon

George Lakoff (1987), Paul Churchland (1995), and Amos Tversky and Daniel

Kahneman (1974), we have sufficient results from cognitive research to

undermine the moral law conception, as well as the autonomous and rational

premises that support it.

There is an impressive quantity of research on the nature of concepts and

categorization. Abstract concepts have been demonstrated to possess a “fuzzy”

nature, by having no clear boundaries and a gradient structure. Thus are

concepts ambiguous at best, and our categorizations under such concepts does

not obey a classical (featural, definitional) semantic theory, but rather some form

of prototype- or exemplar- based processing (Rosch 1978, Smith & Medin 1999).

Moreover, research specifically about moral concepts has shown that moral

prototypes or other non classic conceptual structures result from a radial

1

construction, by exposition to metaphorical and functional relations to other

category members. As if it was not enough, those abstract moral concepts being

defined by sets of metaphors, they exhibit inconsistencies and contradictions

among themselves. As Johnson emphasizes, “… if our fundamental moral

concepts are defined by multiple and possibly inconsistent conceptual

metaphors, then the literalist picture of moral thinking […] cannot be correct."

To complicate things further, it also follows that our reasoning about moral

issues can not be deductive (with exceptions of simplicity in the correspondence

between a situation that would follow literally from a line of moral reasoning).

Moral reasoning is therefore an exploration of correspondence between

metaphorical extensions of prototypical moral situations to atypical ones.

Johnson notes that it does not commit us to the exclusion of moral principles, but

theses principles are not univocal and absolute, they are “idealized” strategies

modulated by experience and categorization via prototypical concepts of

metaphorical structure.

One line of cognitive research has demonstrated that we frame our

reasoning. Our concepts and the relations they hold among themselves are

context-dependent. A moral problem in a situation has many different outlooks

and solutions depending on prior or subordinate clauses, by analogy to research

on decision-making and problem solving in cognitive sciences. Frame

dependence is incompatible with moral law theories and moral fundamentalism.

As a corollary source of reinforcement for the frame-dependent view, there is a

significant body of work that questions the ability of rational agents to make

“good probabilistic judgments of the sort required by an economic conception of

rationality…” writes Johnson. The subject matter here is psychological work on

the use of heuristics and models of decision-making where risk is involved. As

Johnson notes, moral theory draws upon a conception of rationality that surfaced

in classic economics, where agents use their reason, in light of available

information, in an optimal way, to assess their interests and the consequences of

those rational choices.

1

This conception has been proven time and again to be quite unrealistic

psychologically, and the sciences of economics have ruled it out a long time ago,

always eager to borrow on empirical research to further their ends. It is

commonly held nowadays that real agents (as opposed to ideal agents) have

biases and limitations, or again constraints, on their so-called rational choices.

Among the relevant empirical results, people have been shown to be strongly

influenced by the framing of risk assessment, being risk-averse in deliberation

about gains and risk-seeking when losses are concerned, without regards to the

actual objective calculation in terms of optimality. Principles such as satisficing,

as opposed to optimality, and heuristics use, as opposed to maximum

information-oriented decisions, have undermined and replaced the classical

rational perspective in economics and psychology, and ethics should know better

than to stick with ghosts.

On emotions

Dichotomies abound where morals are concerned. An important side

effect of rationalism has been to relegate emotions to a secondary role in the

matter of moral judgment, but prior to Kant, Hume posited morality as an

essentially emotional phenomenon. In Hume’s view, moral judgments were

based on moral sentiments, because it was passions that influenced actions,

whereas reason purported to make judgments about truth (Kant held an

asymmetrical view, where emotions are matter of subjectivity and individuality,

hence incapable of discovering universal principles, the works of pure reason

alone). Hume’s legacy is one where descriptive claims are left out of ethics, since

moral judgments are expressions of emotions.

Johnson condemns the dichotomy as an assumption of radical mutual

exclusion that is again psychologically unrealistic, thus undermining ethical views

on concepts such as motivation and affective judgment. Following Antonio

Damasio’s research in cognitive neurosciences (1994), it had been demonstrated

that practical reasoning is directly dependent on emotional and motivational

1

states, in complex physiologically and neurologically constrained processes.

Research results on cortical lesions and neuropathological afflictions have shown

that reason and emotions are interdependent in the mind’s endeavors to achieve

judgments of practical and moral propensity. A very important empirical matter

here is the demonstration by neurosciences that reason is directly dependent on

its embodiment, a radical turn in the study of rationality from a cognitive direction.

On development of the self and morality

Research on empathy in developmental cognitive science as put forward

the idea that our ability to relate to others through emotional dispositions is the

source of our moral judgments and behaviors. The development of the self is

dependent on socialization, and in a mutually reinforcing interdependence, the

development of our attitudes towards others are dependent on our empathy

(Goldman, 1993). Also worth noting is the ontogeny of infant-parent (particularly

mothers) dynamics in what has been called affect attunement, which is

considered as an essential step in the development of empathy, altruism and

overall socialization. Direct experience of moral situations reinforce the sense of

empathy in ontogeny, from children confronting distress, to the abstraction and

generalization of imagined suffering, to extrapolation into the realm of social

conduct and ethical and judicial tenets.

From Piaget (1932) to Kohlberg (1981), classical views on moral ontogeny

has corroborated the rationalist preconception of moral law, viewing moral

reasoning under the dominion of rationality as the natural end-state of

development of individuals. Flanagan and Johnson argue against the relevance

and realism of the examination standards of such early cognitive-developmental

support to a rationalist conception. Among other concerns, it is not obvious that

people ever develop the so-called later and higher stages of moral thought in

general, since those stages have a highly abstract and formal definition that may

fit the methodology but not the reality. Caeteris paribus, it is the very definitions

of morality and rationality as they are preconceived that influenced the

2

examination standards of those early studies, thus biasing a claim that was

already built-in the method.

Again, classical views on rational agents as being autonomous and

without constraints have been challenged. For one thing, even very modern

rationalist views of morality, such as Rawls’ (1971) and Nozick’s (1974) not quite

compatible views of the political, economical and moral spheres of activity,

presuppose moral agency as a priori and autonomous of actions and

interrelations, thus following kantian rational and free egotism as opposed to a

realistic developmental outlook on the emergence of morality. Also,

developmental concerns have a tendency towards ethical cultural relativism

(Shweder, 1991), underlining through cross-cultural studies that moral ontogeny

is also dependent on exposure to values already in place in the social theatre.

On conceptual schemes used in moral philosophy

A finishing consideration of cognitive inspiration to our subject matter

concerns whether the specifications of what ethics hold as moral orientations are

realistic or not, or in other words whether moral archetypes are over-specified or

under-specified by the argumentation taken in account in moral deliberation. For

example, studies on moral agency from a distinction between genders have a

tendency to segregate a morality of rights and justice from a morality of

responsibilities and care, respectively attributing the former tendency to men and

the later to women, as learned from patriarchal and matriarchal stereotypes that

permeate our culture. Whether or not gender-based studies hold any significant

and heuristic results for the sake of our understanding of ethics is debatable, and

authors are very prudent in their comments with respect to the issue. But the

point raised by Flanagan and Johnson is that we may be too quick to

discriminate moral orientations such as above, and models of ethical inquiry from

even an empirical perspective are biased towards simplicity and strong, radical

segregations among concepts and categories. “… moral personality is, in the

2

end, too variegated and multipurpose to be analyzable in terms of a simple two-

orientation scheme – even blended together” writes Flanagan.

6

Conclusions

Much has been said above on the very large issues of ethical relativism

and how to deal with classical preconceptions of epistemological issues on moral

thought. But in both ways of dealing with a vast area of knowledge and pointing

out to specific issues within them, I have left out a considerable wealth of

questions and answers. My claims can be summed up in the following

paragraphs.

Firstly, I have tried to demonstrate that the concept of relativism as

commonly held is misinterpreted and misconstrued, and does not doom us with

paradoxical or regressive arguments that should play in favor of absolutism. As

such, relativism and absolutism may be but two ends on a spectrum, not an

either/or dichotomy. (Non archaic and non radical) relativism is compatible with a

practical realism, and does not entail that there should be no moral truths,

whereas absolutism rests on unjustifiable metaphysical and epistemological

positions.

Secondly, the issue of an irreconcilable normativity and descriptiveness

brought upon by seemingly absolutist claims of different lineages need not worry

us for much of the same reasons drawn from epistemology and metaphysics.

The possibility of autonomous ethics has been thus demonstrated (time and

again) to be nigh impossible, as with any autonomous research area concerning

the sciences of the living anyhow. The very vocabulary inherited by moral

thought has to keep up with contemporary epistemology and metaphysics, as in

the departure from a criterial view of our knowledge of the world to embrace a

realism of constraints on this knowledge.

2

Thirdly, much of ethics having to do with reason, concepts and agency,

which are the domain of empirical cognitive sciences, we are naturally led to

espouse a dependence of the former area of study on the latter. Not only does it

concur with our previous intuition of the well founded adoption of a correct

relativistic attitude towards knowledge, it eradicates the false pretenses of

autonomous and absolutist ethics, along with the unsupportable claims on the

necessity of untainted normativity by descriptive concerns.

Lastly, the pertinence of ethics is in no way questioned here. What was

dealt with here is about the relativity of ethics to its empirical foundations, of (at

least) cognitive importance. It is a reassessment of ethics’ epistemological and

ontological commitments. This relativity of ethics does not portray ethics as a

failure or ill-fated enterprise, it informs it and constrains it in more realistic ways.

What we might try to achieve is a non-reductive naturalistic realism of ethics,

much in the same way that cognitive sciences resist reductionism and aim to a

rigorously scientific treatment of what might be said to be true about human

mind, behavior and agency.

Future work may go even further than a psychological and cognitive

endeavor to rest ethics in scientific realism. I believe it to be possible to shed

some light on some real and relevant constraints on moral thought and action

within the realms of even lower levels of scientific complexity, from evolutionary

biology down to dynamic systems theory.

“A morality which cannot be revised as new discoveries about the mind are made

known is a dead morality incapable of meeting the kinds of change that are part

of human existence.”

Mark L. Johnson

1998

2

References

- Audi, Robert, 1998. Epistemology: a contemporary introduction to the

theory of knowledge. Routledge, London

- Axelrod, Robert, 1984. The evolution of cooperation. Basic Books, New

York

- Bénatouïl, Thomas, 1997. Le scepticisme. Flammarion, Paris

- Benedict, Ruth, 1934 (ed. 1989). Patterns of culture. Mariner Books,

Boston

- Carnap, Rudolf, 1928. Der logische Aufbau der Welt. Felix Meiner Verlag,

Leipzig

- Church, Alonso, 1936. An unsolvable problem in elementary number

theory, in American Journal of Mathematics, vol. 58, pp. 345-363

- Churchland, Paul, 1995. The engine of reason, the seat of the soul: a

philosophical journey into the brain. MIT Press, Cambridge

- Coveney, P., Highfield, R., 1995. Frontiers of complexity. Ballantine

Books, New York

- Damasio, Antonio, 1994. Descartes’ error: emotion, reason, and the

human brain. Grosset/Putnam, New York

- Dawkins, Richard, 1986 (ed. 1996). The blind watchmaker: why the

evidence of evolution reveals a universe without design. W.W. Norton &

Co., New York

- Dennett, Daniel, 1995. Darwin’s dangerous idea: evolution and the

meanings of life. Touchstone, New York

- Dennett, Daniel, 2003. Freedom evolves. Viking Press, London

2

- Dewey, John, 1922. Human nature and conduct. In Boydston, J., 1983.

The middle works of John Dewey. Southern Illinois University Press,

Carbondale, Illinois

- Dummett, Michael, 1993 (ed. 1996). The seas of language. Oxford

University Press, Oxford

- Einstein, Albert, 1918 (ed. 1995) Relativity: the special and the general

theory. Three Rivers Press, New York

- Flanagan, Owen, 1991. Varieties of moral personality: ethics and

psychological realism. Harvard University Press, Cambridge

- Gödel, Kurt, 1931. Über formal unentscheidbare Sätze der Principia

Mathematica und verwandter Systeme, in Monatshefte für Mathematik

und Physik, vol. 38

- Gleick, James, 1988. Chaos. Sphere, London

- Goldman, A., 1993. Ethics and cognitive science. In Ethics, no. 103, pp.

337-360

- Harman, Gilbert, 1985. Is there a single true morality? In Copp, D.,

Zimmerman, D., 1985. Morality, reason and truth. Rowman & Allanheld,

Totowa, New Jersey

- Hume, David, 1739 (ed. 1986). Treatise of human nature. Viking Press,

London

- Johnson, Mark, 1993. Moral Imagination: implications of cognitive science

for ethics. University of Chicago Press, Chicago

- Johnson, Mark, 1998. Ethics. In Bechtel, W., Graham, G. 1999. A

companion to cognitive science. Blackwell, Oxford

- Kant, Immanuel, 1797 (ed. 1959). Foundations of the metaphysics of

morals. MacMillan Publishing Company, New York

- Kauffman, Stuart, 1993. The origins of order: self-organization and

selection in evolution. Oxford University Press, Oxford

- Kohlberg, Lawrence, 1981. Essays on moral development, vol. 1: the

philosophy of moral development. Harper & Row, San Francisco

2

- Lakoff, G., 1987. Women, fire, and dangerous things: what categories

reveal about the mind. University of Chicago Press, Chicago

- Lakoff, G., 1996. Moral politics: what conservatives know that liberals

don’t. University of Chicago Press, Chicago

- Lewin, Roger, 1993. Complexity. Dent, London

- Millikan, Ruth, 1987. Language, thought, and other biological categories.

MIT Press, Cambridge

- Mill, John S., 1861 (ed. 1998). Utilitarianism. Oxford Press, Oxford

- Moore, G. E., 1903 (ed. 1988). Principia Ethica. Prometheus Books,

Cambridge University Press, Cambridge

- Nash, John Forbes Jr., 1997. Essays on game theory. Edward Elgar Pub.,

Vermont

- Newell, Allen, 1990. Unified Theories of Cognition. Harvard University

Press, Cambridge

- Nozick, Robert, 1974 (ed. 1977). Anarchy, state, and utopia. Basic Books,

New York

- Piaget, Jean, 1932 (ed. 1997). The moral judgment of the child. Free

Press Paperbacks, New York

- Quine, W.V.O., original 1951, 1953 (ed. 1961). Two dogmas of

empiricism, in From a logical point of view. Harvard University Press,

Cambridge

- Quine, W.V.O., 1990 (ed. 1992). The pursuit of truth. Harvard University

Press, Cambridge

- Rawls, John, 1971 (ed. 1999). A theory of justice. Belknap Press, Harvard

University Press, Cambridge

- Rosch, Eleanor, 1978. Principles of Categorization. In Rosch, E., Lloyd,

B., 1978. Cognition and Categorization. Lawrence Erlbaum & co.,

Hillsdale, New Jersey

- Shweder, Richard, 1991. Thinking through cultures: expeditions in cultural

psychology. Harvard University Press, Cambridge

- Simon, Herbert, 1956 (ed. 1987). Models of man. Garland Pub., New York

2

- Smith, E., Medin, D., 1999. The examplar view. In Margolis, E., Laurence,

S., 1999. Concepts: core readings. MIT Press, Cambridge

- Stewart, Ian, 1990. Does God play dice ? Blackwell, Oxford

- Tversky, A., Kahneman, D., 1974. Judgment under uncertainty: heuristics

and biases. In Science, no. 185, pp. 1124-1131

2